The aim of this proposal is to evaluate the role of deoxycytidine (dCyd) as a growth protective agent for normal myeloid progenitor cells in antileukemic regimens consisting of dCyd analogs, or other pyrimidine antagonists, and high dose dCyd. This effort will attempt to uncover the biochemical mechanisms underlying a novel observation that supraphysiologic concentrations of this naturally occurring nucleoside selectively stimulates the in vitro growth of normal vs leukemic myeloid progenitor cells and preferentially protects the former from the lethal actions of certain pyrimidine antagonists, particularly 1-B- D-arabinofuranosylcytosine (Ara-C). Biochemical studies will be performed comparing the intracellular metabolism of high dose dCyd in freshly procured normal human bone marrow mononuclear cells and human leukemic myeloblasts. Specifically, we will study the intracellular degradation of dCyd and dCyd monophosphate (dCMP) and the accumulation of intracellular dCyd triphosphate (dCTP) pools (utilizing DNA polymerase assay). Similarly, both in normal and leukemic cells exposed to high dose dCyd/Ara-C regimens, we will carefully examine the accumulation of Ara- CTP; Ara-CTP to dCTP pool ratios (utilizing high pressure liquid chromatography methods); deamination of dCyd (dCMP) and Ara- C (Ara-CMP) by nucleoside and dCMP deaminase; incorporation of Ara-C into newly synthesized DNA; and molecular phenomenon such as the proportion of Ara-C residues at 3'-terminus of DNA strand to uncover possible explanations for the selective antileukemic effect of high dose dCyd/Ara-C regimens. In biologic studies, utilizing soft agar cloning methods, we will identify high dose dCyd and dCyd analog (e.g. 5-aza-Ara-C and 5- Aza-dCyd) dose relationship and schedules which exert lethal effects toward leukemic blast progenitors (L-CFU) while sparing normal myeloid elements (CFU-GM). The selective ability of dCyd to protect normal vs leukemic cells from other pyrimidine antagonists (e.g. thymidine and 3-deazauridine) will also be examined. Correlations with biochemical studies will be sought in order to test the hypothesis that impaired dCyd nucleotide formation or enhanced degradation in leukemic cells is responsible for diminished antagonism of pyrimidine antagonist mediated cytotoxicity. The ultimate objective of these studies is to generate in vitro data which will establish a rational basis for the design and implementation of clinical trails of high dose dCyd/dCyd analog, or high dose dCyd/pyrimidine antagonist, regimens in the treatment of acute nonlymphocytic leukemia.